Interpretable Diagnosis for Whole-Slide Melanoma Histology Images Using Convolutional Neural Network

At present, deep learning-based medical image diagnosis had achieved high performance in several diseases. However, the black-box nature of the convolutional neural network (CNN) limits their role in diagnosis. In this study, a novel interpretable diagnosis pipeline using the CNN model was proposed....

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Peizhen Xie, Ke Zuo, Jie Liu, Mingliang Chen, Shuang Zhao, Wenjie Kang, Fangfang Li
Formato:	article
Lenguaje:	EN
Publicado:	Hindawi Limited 2021
Materias:	Medicine (General) R5-920 Medical technology R855-855.5
Acceso en línea:	https://doaj.org/article/c8a87f1ace064cb594f76c517c762f96
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:c8a87f1ace064cb594f76c517c762f96
record_format	dspace
spelling	oai:doaj.org-article:c8a87f1ace064cb594f76c517c762f962021-11-15T01:19:52ZInterpretable Diagnosis for Whole-Slide Melanoma Histology Images Using Convolutional Neural Network2040-230910.1155/2021/8396438https://doaj.org/article/c8a87f1ace064cb594f76c517c762f962021-01-01T00:00:00Zhttp://dx.doi.org/10.1155/2021/8396438https://doaj.org/toc/2040-2309At present, deep learning-based medical image diagnosis had achieved high performance in several diseases. However, the black-box nature of the convolutional neural network (CNN) limits their role in diagnosis. In this study, a novel interpretable diagnosis pipeline using the CNN model was proposed. Furthermore, a sizeable melanoma database that contains 841 digital whole-slide images (WSIs) was built to train and evaluate the model. The model achieved strong melanoma classification ability (0.962 areas under the receiver operating characteristic, 0.887 sensitivity, and 0.925 specificity). Moreover, the proposed model outperformed the existing schemes in terms of accuracy that is 20 pathologists (0.933 vs 0.732 accuracy). Finally, the gradient-weighted class activation mapping (Grad-CAM) method was used to show the inner logic of the proposed model and its feasibility to improve diagnosis process in healthcare. The mechanism of feature heat maps which is visualized through a saliency mapping has demonstrated that features learned or extracted by the proposed model are compatible with the accepted pathological features. Conclusively, the proposed model provides a rapid and accurate diagnosis by locating the distinctive features of melanoma to build doctors’ trust in the CNNs’ diagnosis results.Peizhen XieKe ZuoJie LiuMingliang ChenShuang ZhaoWenjie KangFangfang LiHindawi LimitedarticleMedicine (General)R5-920Medical technologyR855-855.5ENJournal of Healthcare Engineering, Vol 2021 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine (General) R5-920 Medical technology R855-855.5
spellingShingle	Medicine (General) R5-920 Medical technology R855-855.5 Peizhen Xie Ke Zuo Jie Liu Mingliang Chen Shuang Zhao Wenjie Kang Fangfang Li Interpretable Diagnosis for Whole-Slide Melanoma Histology Images Using Convolutional Neural Network
description	At present, deep learning-based medical image diagnosis had achieved high performance in several diseases. However, the black-box nature of the convolutional neural network (CNN) limits their role in diagnosis. In this study, a novel interpretable diagnosis pipeline using the CNN model was proposed. Furthermore, a sizeable melanoma database that contains 841 digital whole-slide images (WSIs) was built to train and evaluate the model. The model achieved strong melanoma classification ability (0.962 areas under the receiver operating characteristic, 0.887 sensitivity, and 0.925 specificity). Moreover, the proposed model outperformed the existing schemes in terms of accuracy that is 20 pathologists (0.933 vs 0.732 accuracy). Finally, the gradient-weighted class activation mapping (Grad-CAM) method was used to show the inner logic of the proposed model and its feasibility to improve diagnosis process in healthcare. The mechanism of feature heat maps which is visualized through a saliency mapping has demonstrated that features learned or extracted by the proposed model are compatible with the accepted pathological features. Conclusively, the proposed model provides a rapid and accurate diagnosis by locating the distinctive features of melanoma to build doctors’ trust in the CNNs’ diagnosis results.
format	article
author	Peizhen Xie Ke Zuo Jie Liu Mingliang Chen Shuang Zhao Wenjie Kang Fangfang Li
author_facet	Peizhen Xie Ke Zuo Jie Liu Mingliang Chen Shuang Zhao Wenjie Kang Fangfang Li
author_sort	Peizhen Xie
title	Interpretable Diagnosis for Whole-Slide Melanoma Histology Images Using Convolutional Neural Network
title_short	Interpretable Diagnosis for Whole-Slide Melanoma Histology Images Using Convolutional Neural Network
title_full	Interpretable Diagnosis for Whole-Slide Melanoma Histology Images Using Convolutional Neural Network
title_fullStr	Interpretable Diagnosis for Whole-Slide Melanoma Histology Images Using Convolutional Neural Network
title_full_unstemmed	Interpretable Diagnosis for Whole-Slide Melanoma Histology Images Using Convolutional Neural Network
title_sort	interpretable diagnosis for whole-slide melanoma histology images using convolutional neural network
publisher	Hindawi Limited
publishDate	2021
url	https://doaj.org/article/c8a87f1ace064cb594f76c517c762f96
work_keys_str_mv	AT peizhenxie interpretablediagnosisforwholeslidemelanomahistologyimagesusingconvolutionalneuralnetwork AT kezuo interpretablediagnosisforwholeslidemelanomahistologyimagesusingconvolutionalneuralnetwork AT jieliu interpretablediagnosisforwholeslidemelanomahistologyimagesusingconvolutionalneuralnetwork AT mingliangchen interpretablediagnosisforwholeslidemelanomahistologyimagesusingconvolutionalneuralnetwork AT shuangzhao interpretablediagnosisforwholeslidemelanomahistologyimagesusingconvolutionalneuralnetwork AT wenjiekang interpretablediagnosisforwholeslidemelanomahistologyimagesusingconvolutionalneuralnetwork AT fangfangli interpretablediagnosisforwholeslidemelanomahistologyimagesusingconvolutionalneuralnetwork
_version_	1718428948448149504

Interpretable Diagnosis for Whole-Slide Melanoma Histology Images Using Convolutional Neural Network

Ejemplares similares